The people reading the research so you don't have to skim it
Four writers, one shared habit: check the public data before repeating a claim about speed.
Maren Kessler
Research Editor
Maren spends most weeks inside Google's published web performance documentation, the Chrome team's engineering posts, and academic papers on user attention. Her job is less about discovery and more about translation: taking a dense technical explanation and finding the plain-language version that doesn't lose the caveats along the way.
- Reads every new web.dev performance post within a week of publication
- Maintains the internal glossary of terms used across the blog
- Flags when an article overstates what the source actually said
Devon Ualani
Data Analyst
Devon works with the publicly queryable Chrome User Experience Report and periodic HTTP Archive crawls, mostly to test whether widely repeated speed claims still hold up once you look at the underlying distributions rather than the averages. Averages hide a lot. Devon's job is to notice what they're hiding.
- Builds small, checkable queries against public CrUX datasets
- Cross-references claims from other publications before we repeat them
- Prefers a boring, accurate finding over an exciting, shaky one
Priya Nadkarni
Walkthroughs Writer
Priya writes the screenshot walkthroughs, which means she spends a strange amount of time thinking about how a Lighthouse report looks to someone who has never opened developer tools before. She tests every explanation on people outside the industry before it gets published.
- Runs Lighthouse against a rotating set of public pages for examples
- Writes for readers with zero assumed technical background
- Revises any explanation that a non-technical reader stumbles on
Samuel Ortiz
Tools Curator
Samuel maintains the free tools page, which sounds simple until a tool changes its interface or quietly starts gating a feature behind a paywall. He rechecks every listed tool on a regular basis and updates the notes when something shifts.
- Verifies every tool on the site is still free to use as described
- Notes the practical limits of each tool, not just the features
- Removes anything that stops matching how it's described here
Behind the scenes
How an article actually gets written
01
Start with the primary source
Every article begins with Google's own documentation or a public dataset, not a summary of a summary. If we can't find the original source, the claim doesn't get published.
02
Check it against the data
Devon runs a check against CrUX or HTTP Archive where it's applicable, mostly to see whether a claim generalizes or only applies to a narrow slice of sites.
03
Write it for someone who isn't a developer
Priya or Maren draft the piece assuming no prior technical knowledge, and define every acronym the first time it appears.
04
Read it back out loud
A strange but effective habit. If a sentence sounds like a sales pitch when read aloud, it gets rewritten until it sounds like a note from a colleague instead.
"We're not trying to convince anyone their site is slow. We're trying to explain why speed and behavior are connected, and let people decide what that means for their own page."
Maren Kessler, Research Editor
What this site doesn't do
This is a publication, not a vendor. We don't sell speed audits, optimization retainers, or consulting hours. When an article mentions a tool or a technique, it's because we used it to understand a public dataset or a Lighthouse report, not because we're recommending a purchase. If you're looking for someone to fix a slow page, this blog can help you understand the report. It won't send you an invoice.